Faceted search, that is, search wherein search results are filtered based upon aspects or facets that are particular to the query being searched, has gained popularity as a way of presenting search results so as to increase user engagement therewith. Objects and/or subjects of a query often have a large number of facets on which they can be filtered. Thus, determining on which facets to filter search results such that user engagement is maximized is important to insuring that the full benefits of faceted search are realized. Generally, this is a highly manual process wherein a particular query is received, various facet-filters decided upon by knowledge workers or domain experts are applied to a set of results for the particular query, the results are presented to various users in accordance therewith, and user engagement with the arrangement of results is monitored. Based upon manual analysis of the user engagement data, a single, most successful arrangement is decided upon and shown to all future users upon input of the particular query.